Our success in producing large quantities of crystallization-grade proteins led to a small-scale structural genomics project aiming to solve the three-dimensional structures of proteins involved in Type III secretion in Yersinia pestis, the causative agent of plague. Because the Type III secretion system (T3SS) is essential for virulence, the resulting structural information could be used to develop effective countermeasures for this potential agent of bioterrorism. We have already solved 15 novel structures and are in the process of solving more of them, including several protein-protein complexes. However, the structural genomics aspect of this project is being phased out for a variety of reasons.Our current focus has shifted o the process of structure-assisted drug development. One of the cytotoxic effector proteins that Yersinia injects into mammalian cells via the T3SS, YopH, is a potent eukaryotic-like protein tyrosine phosphatase (PTPase). YopH dephosphorylates several proteins associated with the focal adhesion in eukaryotic cells, thereby enabling the bacterium to avoid phagocytosis and destruction by macrophages. In collaboration with Dr. Terrence Burke Jr. (Laboratory of Medicinal Chemistry, CCR) and Dr. Robert Ulrich (USAMRIID), we have recently developed a highly specific and potent inhibitor of YopH that is non promiscuous, nontoxic to mammalian cells and very effective at killing the bacterium in macrophages. Plans are underway to test this compound in animal models of plague infection.The nsp2 protease encoded by Venezuelan Equine Encephalitis virus (VEEV) is a potential target for therapeutic antiviral agents. We crystallized the catalytic domain of the protease and determined its structure. However, for a variety of reasons, the crystals we obtained were less than ideal for a drug development project. Recently we have engineered a new crystal form of the enzyme, by surface entropy reduction mutagenesis, that is much better suited for drug development efforts and determined its structure at a resolution of 1.3 Angstroms. We are currently performing in silico screening to identify lead molecules for co-crystallization with the enzyme and further optimization.